Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
32 changes: 31 additions & 1 deletion marker/builders/layout.py
Original file line number Diff line number Diff line change
@@ -1,3 +1,4 @@
import json
from typing import Annotated, List

from surya.layout import LayoutPredictor
Expand Down Expand Up @@ -27,6 +28,10 @@ class LayoutBuilder(BaseBuilder):
str,
"Skip layout and force every page to be treated as a specific block type.",
] = None
force_layout_str: Annotated[
str,
"JSON string of layout for each page. Useful for using an altenative layout model.",
] = None
disable_tqdm: Annotated[
bool,
"Disable tqdm progress bars.",
Expand All @@ -49,7 +54,9 @@ def __init__(self, layout_model: LayoutPredictor, config=None):
super().__init__(config)

def __call__(self, document: Document, provider: PdfProvider):
if self.force_layout_block is not None:
if self.force_layout_str is not None:
layout_results = self.forced_layout_from_str(document.pages)
elif self.force_layout_block is not None:
# Assign the full content of every page to a single layout type
layout_results = self.forced_layout(document.pages)
else:
Expand Down Expand Up @@ -91,6 +98,29 @@ def surya_layout(self, pages: List[PageGroup]) -> List[LayoutResult]:
)
return layout_results

def forced_layout_from_str(self, pages: List[PageGroup]) -> List[LayoutResult]:
layout_json = json.loads(self.force_layout_str)
assert len(layout_json) == len(pages)
layout_results = []
for page, forced_layout in zip(pages, layout_json):
for i, layout_block in enumerate(forced_layout):
layout_results.append(
LayoutResult(
image_bbox=page.polygon.bbox,
bboxes=[
LayoutBox(
label=layout_block["label"],
position=i,
top_k={layout_block["label"]: 1},
polygon=PolygonBox.from_bbox(layout_block["bbox"]).polygon,
),
],
sliced=False,
)
)

return layout_results

def expand_layout_blocks(self, document: Document):
for page in document.pages:
# Collect all blocks on this page as PolygonBox for easy access
Expand Down